ANALYSIS OF QS-21 INTERACTIONS USING MOLECULAR DYNAMICS SIMULATIONS

Saponins are glycosides of plant origin that have been extensively studied for their adjuvant potential—that is, their ability to stimulate the immune system. This property makes them valuable in the development of medicines and vaccines for both human and veterinary use. In this context, Quillaja saponaria, a plant native to South America, stands out as the source of QS-21, a nontoxic and abundant extract that is the focus of this research. QS-21 has a complex molecular structure, composed of a triterpenoid backbone linked to sugar chains. Our research group observed, through small-angle X-ray scattering (SAXS), the formation of elongated micelles in solution at concentrations around 5%, which become spherical at 8% (Pedebos et al., Molecules, 2014, 3744). This work aims to study the intramolecular interactions to understand and validate the structural transformations in the system. To achieve this, all-atom molecular dynamics simulations of QS-21 in aqueous solution are being conducted using GPU-supported GROMACS. A series of simulations has been proposed, varying QS-21 concentrations under NPT conditions. Pressure and temperature are maintained constant at 1 bar and 298 K. From the resulting trajectories, structures formed by QS-21 are being evaluated, including aggregate size and dynamic behavior. The structuring of the solvent around the saponin is also being analyzed using the radial distribution function (RDF), minimum-distance RDF, Kirkwood-Buff integrals, solvent coordination numbers, and hydrogen bonds (saponin–saponin, water–water, and saponin–water). So far, the simulation for a 3% QS-21 system has been completed. After 250 ns, all 20 molecules compacted into a single aggregate with a radius of gyration of 4.08 nm, slightly larger than the micelle radius obtained via SAXS. RDFs and Kirkwood-Buff integrals indicated strong solute aggregation. Additionally, Kirkwood-Buff theory allows the calculation of thermodynamic parameters of the system. Two remaining simulations, at 5% and 8% concentrations, are prepared and await computational availability. Once completed, the parameters will be comparatively analyzed to understand the molecular organization as a function of concentration.